Content Based Image Retrieval system using Wavelet Transformation and multiple input multiple task Deep Autoencoder

Xiangyuan Zhao, Brian Nutter

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Scopus citations

Abstract

In this paper, we propose an algorithm for a Content Based Image Retrieval (CBIR) system based on Wavelet Transformation and Deep Autoencoder (DAE). For the proposed algorithm, the image is first processed by wavelet transform and decomposed into wavelet coefficients. The wavelet coefficients then become the input for a multiple input multiple task deep autoencoder (MIMT-DAE). In our design, only the approximation coefficients (CA) and diagonal detail coefficients (CD) are used. The result of retrieval performance is tested on the MNIST handwriting data base. The testing results show that the combination of wavelet transformation and MIMT-DAE increases the performance of image retrieval for shape and texture compared to a traditional single input single task deep autoencoder with far fewer training parameters required.

Original languageEnglish
Title of host publication2016 IEEE Southwest Symposium on Image Analysis and Interpretation, SSIAI 2016 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages97-100
Number of pages4
ISBN (Electronic)9781467399197
DOIs
StatePublished - Apr 25 2016
EventIEEE Southwest Symposium on Image Analysis and Interpretation, SSIAI 2016 - Santa Fe, United States
Duration: Mar 6 2016Mar 8 2016

Publication series

NameProceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation
Volume2016-April

Conference

ConferenceIEEE Southwest Symposium on Image Analysis and Interpretation, SSIAI 2016
CountryUnited States
CitySanta Fe
Period03/6/1603/8/16

Keywords

  • CBIR
  • Deep Neural Network
  • Multiple Input Multiple Task Deep Autoencoder
  • Wavelet Transformation

Fingerprint Dive into the research topics of 'Content Based Image Retrieval system using Wavelet Transformation and multiple input multiple task Deep Autoencoder'. Together they form a unique fingerprint.

  • Cite this

    Zhao, X., & Nutter, B. (2016). Content Based Image Retrieval system using Wavelet Transformation and multiple input multiple task Deep Autoencoder. In 2016 IEEE Southwest Symposium on Image Analysis and Interpretation, SSIAI 2016 - Proceedings (pp. 97-100). [7459184] (Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation; Vol. 2016-April). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SSIAI.2016.7459184